3 perlfaq5 - Files and Formats ($Revision: 3606 $)
7 This section deals with I/O and the "f" issues: filehandles, flushing,
10 =head2 How do I flush/unbuffer an output filehandle? Why must I do this?
11 X<flush> X<buffer> X<unbuffer> X<autoflush>
13 Perl does not support truly unbuffered output (except
14 insofar as you can C<syswrite(OUT, $char, 1)>), although it
15 does support is "command buffering", in which a physical
16 write is performed after every output command.
18 The C standard I/O library (stdio) normally buffers
19 characters sent to devices so that there isn't a system call
20 for each byte. In most stdio implementations, the type of
21 output buffering and the size of the buffer varies according
22 to the type of device. Perl's print() and write() functions
23 normally buffer output, while syswrite() bypasses buffering
26 If you want your output to be sent immediately when you
27 execute print() or write() (for instance, for some network
28 protocols), you must set the handle's autoflush flag. This
29 flag is the Perl variable $| and when it is set to a true
30 value, Perl will flush the handle's buffer after each
31 print() or write(). Setting $| affects buffering only for
32 the currently selected default file handle. You choose this
33 handle with the one argument select() call (see
34 L<perlvar/$E<verbar>> and L<perlfunc/select>).
36 Use select() to choose the desired handle, then set its
37 per-filehandle variables.
39 $old_fh = select(OUTPUT_HANDLE);
43 Some idioms can handle this in a single statement:
45 select((select(OUTPUT_HANDLE), $| = 1)[0]);
47 $| = 1, select $_ for select OUTPUT_HANDLE;
49 Some modules offer object-oriented access to handles and their
50 variables, although they may be overkill if this is the only
51 thing you do with them. You can use IO::Handle:
54 open(DEV, ">/dev/printer"); # but is this?
59 use IO::Socket; # this one is kinda a pipe?
60 my $sock = IO::Socket::INET->new( 'www.example.com:80' );
64 =head2 How do I change one line in a file/delete a line in a file/insert a line in the middle of a file/append to the beginning of a file?
67 Use the Tie::File module, which is included in the standard
68 distribution since Perl 5.8.0.
70 =head2 How do I count the number of lines in a file?
71 X<file, counting lines> X<lines> X<line>
73 One fairly efficient way is to count newlines in the file. The
74 following program uses a feature of tr///, as documented in L<perlop>.
75 If your text file doesn't end with a newline, then it's not really a
76 proper text file, so this may report one fewer line than you expect.
79 open(FILE, $filename) or die "Can't open `$filename': $!";
80 while (sysread FILE, $buffer, 4096) {
81 $lines += ($buffer =~ tr/\n//);
85 This assumes no funny games with newline translations.
87 =head2 How can I use Perl's C<-i> option from within a program?
90 C<-i> sets the value of Perl's C<$^I> variable, which in turn affects
91 the behavior of C<< <> >>; see L<perlrun> for more details. By
92 modifying the appropriate variables directly, you can get the same
93 behavior within a larger program. For example:
97 local($^I, @ARGV) = ('.orig', glob("*.c"));
100 print "This line should appear at the top of each file\n";
102 s/\b(p)earl\b/${1}erl/i; # Correct typos, preserving case
104 close ARGV if eof; # Reset $.
107 # $^I and @ARGV return to their old values here
109 This block modifies all the C<.c> files in the current directory,
110 leaving a backup of the original data from each file in a new
113 =head2 How can I copy a file?
114 X<copy> X<file, copy>
116 (contributed by brian d foy)
118 Use the File::Copy module. It comes with Perl and can do a
119 true copy across file systems, and it does its magic in
124 copy( $original, $new_copy ) or die "Copy failed: $!";
126 If you can't use File::Copy, you'll have to do the work yourself:
127 open the original file, open the destination file, then print
128 to the destination file as you read the original.
130 =head2 How do I make a temporary file name?
133 If you don't need to know the name of the file, you can use C<open()>
134 with C<undef> in place of the file name. The C<open()> function
135 creates an anonymous temporary file.
137 open my $tmp, '+>', undef or die $!;
139 Otherwise, you can use the File::Temp module.
141 use File::Temp qw/ tempfile tempdir /;
143 $dir = tempdir( CLEANUP => 1 );
144 ($fh, $filename) = tempfile( DIR => $dir );
146 # or if you don't need to know the filename
148 $fh = tempfile( DIR => $dir );
150 The File::Temp has been a standard module since Perl 5.6.1. If you
151 don't have a modern enough Perl installed, use the C<new_tmpfile>
152 class method from the IO::File module to get a filehandle opened for
153 reading and writing. Use it if you don't need to know the file's name:
156 $fh = IO::File->new_tmpfile()
157 or die "Unable to make new temporary file: $!";
159 If you're committed to creating a temporary file by hand, use the
160 process ID and/or the current time-value. If you need to have many
161 temporary files in one process, use a counter:
165 my $temp_dir = -d '/tmp' ? '/tmp' : $ENV{TMPDIR} || $ENV{TEMP};
166 my $base_name = sprintf("%s/%d-%d-0000", $temp_dir, $$, time());
171 until (defined(fileno(FH)) || $count++ > 100) {
172 $base_name =~ s/-(\d+)$/"-" . (1 + $1)/e;
173 # O_EXCL is required for security reasons.
174 sysopen(FH, $base_name, O_WRONLY|O_EXCL|O_CREAT);
177 if (defined(fileno(FH))
178 return (*FH, $base_name);
186 =head2 How can I manipulate fixed-record-length files?
187 X<fixed-length> X<file, fixed-length records>
189 The most efficient way is using L<pack()|perlfunc/"pack"> and
190 L<unpack()|perlfunc/"unpack">. This is faster than using
191 L<substr()|perlfunc/"substr"> when taking many, many strings. It is
192 slower for just a few.
194 Here is a sample chunk of code to break up and put back together again
195 some fixed-format input lines, in this case from the output of a normal,
199 # 15158 p5 T 0:00 perl /home/tchrist/scripts/now-what
200 my $PS_T = 'A6 A4 A7 A5 A*';
201 open my $ps, '-|', 'ps';
203 my @fields = qw( pid tt stat time command );
206 @process{@fields} = unpack($PS_T, $_);
207 for my $field ( @fields ) {
208 print "$field: <$process{$field}>\n";
210 print 'line=', pack($PS_T, @process{@fields} ), "\n";
213 We've used a hash slice in order to easily handle the fields of each row.
214 Storing the keys in an array means it's easy to operate on them as a
215 group or loop over them with for. It also avoids polluting the program
216 with global variables and using symbolic references.
218 =head2 How can I make a filehandle local to a subroutine? How do I pass filehandles between subroutines? How do I make an array of filehandles?
219 X<filehandle, local> X<filehandle, passing> X<filehandle, reference>
221 As of perl5.6, open() autovivifies file and directory handles
222 as references if you pass it an uninitialized scalar variable.
223 You can then pass these references just like any other scalar,
224 and use them in the place of named handles.
226 open my $fh, $file_name;
228 open local $fh, $file_name;
230 print $fh "Hello World!\n";
234 If you like, you can store these filehandles in an array or a hash.
235 If you access them directly, they aren't simple scalars and you
236 need to give C<print> a little help by placing the filehandle
237 reference in braces. Perl can only figure it out on its own when
238 the filehandle reference is a simple scalar.
240 my @fhs = ( $fh1, $fh2, $fh3 );
242 for( $i = 0; $i <= $#fhs; $i++ ) {
243 print {$fhs[$i]} "just another Perl answer, \n";
247 Before perl5.6, you had to deal with various typeglob idioms
248 which you may see in older code.
250 open FILE, "> $filename";
251 process_typeglob( *FILE );
252 process_reference( \*FILE );
254 sub process_typeglob { local *FH = shift; print FH "Typeglob!" }
255 sub process_reference { local $fh = shift; print $fh "Reference!" }
257 If you want to create many anonymous handles, you should
258 check out the Symbol or IO::Handle modules.
260 =head2 How can I use a filehandle indirectly?
261 X<filehandle, indirect>
263 An indirect filehandle is using something other than a symbol
264 in a place that a filehandle is expected. Here are ways
265 to get indirect filehandles:
267 $fh = SOME_FH; # bareword is strict-subs hostile
268 $fh = "SOME_FH"; # strict-refs hostile; same package only
269 $fh = *SOME_FH; # typeglob
270 $fh = \*SOME_FH; # ref to typeglob (bless-able)
271 $fh = *SOME_FH{IO}; # blessed IO::Handle from *SOME_FH typeglob
273 Or, you can use the C<new> method from one of the IO::* modules to
274 create an anonymous filehandle, store that in a scalar variable,
275 and use it as though it were a normal filehandle.
277 use IO::Handle; # 5.004 or higher
278 $fh = IO::Handle->new();
280 Then use any of those as you would a normal filehandle. Anywhere that
281 Perl is expecting a filehandle, an indirect filehandle may be used
282 instead. An indirect filehandle is just a scalar variable that contains
283 a filehandle. Functions like C<print>, C<open>, C<seek>, or
284 the C<< <FH> >> diamond operator will accept either a named filehandle
285 or a scalar variable containing one:
287 ($ifh, $ofh, $efh) = (*STDIN, *STDOUT, *STDERR);
288 print $ofh "Type it: ";
290 print $efh "What was that: $got";
292 If you're passing a filehandle to a function, you can write
293 the function in two ways:
297 print $fh "Sending to indirect filehandle\n";
300 Or it can localize a typeglob and use the filehandle directly:
304 print FH "Sending to localized filehandle\n";
307 Both styles work with either objects or typeglobs of real filehandles.
308 (They might also work with strings under some circumstances, but this
314 In the examples above, we assigned the filehandle to a scalar variable
315 before using it. That is because only simple scalar variables, not
316 expressions or subscripts of hashes or arrays, can be used with
317 built-ins like C<print>, C<printf>, or the diamond operator. Using
318 something other than a simple scalar variable as a filehandle is
319 illegal and won't even compile:
321 @fd = (*STDIN, *STDOUT, *STDERR);
322 print $fd[1] "Type it: "; # WRONG
323 $got = <$fd[0]> # WRONG
324 print $fd[2] "What was that: $got"; # WRONG
326 With C<print> and C<printf>, you get around this by using a block and
327 an expression where you would place the filehandle:
329 print { $fd[1] } "funny stuff\n";
330 printf { $fd[1] } "Pity the poor %x.\n", 3_735_928_559;
331 # Pity the poor deadbeef.
333 That block is a proper block like any other, so you can put more
334 complicated code there. This sends the message out to one of two places:
337 print { $ok ? $fd[1] : $fd[2] } "cat stat $ok\n";
338 print { $fd[ 1+ ($ok || 0) ] } "cat stat $ok\n";
340 This approach of treating C<print> and C<printf> like object methods
341 calls doesn't work for the diamond operator. That's because it's a
342 real operator, not just a function with a comma-less argument. Assuming
343 you've been storing typeglobs in your structure as we did above, you
344 can use the built-in function named C<readline> to read a record just
345 as C<< <> >> does. Given the initialization shown above for @fd, this
346 would work, but only because readline() requires a typeglob. It doesn't
347 work with objects or strings, which might be a bug we haven't fixed yet.
349 $got = readline($fd[0]);
351 Let it be noted that the flakiness of indirect filehandles is not
352 related to whether they're strings, typeglobs, objects, or anything else.
353 It's the syntax of the fundamental operators. Playing the object
354 game doesn't help you at all here.
356 =head2 How can I set up a footer format to be used with write()?
359 There's no builtin way to do this, but L<perlform> has a couple of
360 techniques to make it possible for the intrepid hacker.
362 =head2 How can I write() into a string?
363 X<write, into a string>
365 See L<perlform/"Accessing Formatting Internals"> for an swrite() function.
367 =head2 How can I output my numbers with commas added?
370 (contributed by brian d foy and Benjamin Goldberg)
372 You can use L<Number::Format> to separate places in a number.
373 It handles locale information for those of you who want to insert
374 full stops instead (or anything else that they want to use,
377 This subroutine will add commas to your number:
381 1 while s/^([-+]?\d+)(\d{3})/$1,$2/;
385 This regex from Benjamin Goldberg will add commas to numbers:
387 s/(^[-+]?\d+?(?=(?>(?:\d{3})+)(?!\d))|\G\d{3}(?=\d))/$1,/g;
389 It is easier to see with comments:
392 ^[-+]? # beginning of number.
393 \d+? # first digits before first comma
394 (?= # followed by, (but not included in the match) :
395 (?>(?:\d{3})+) # some positive multiple of three digits.
396 (?!\d) # an *exact* multiple, not x * 3 + 1 or whatever.
399 \G\d{3} # after the last group, get three digits
400 (?=\d) # but they have to have more digits after them.
403 =head2 How can I translate tildes (~) in a filename?
404 X<tilde> X<tilde expansion>
406 Use the <> (glob()) operator, documented in L<perlfunc>. Older
407 versions of Perl require that you have a shell installed that groks
408 tildes. Recent perl versions have this feature built in. The
409 File::KGlob module (available from CPAN) gives more portable glob
412 Within Perl, you may use this directly:
415 ^ ~ # find a leading tilde
417 [^/] # a non-slash character
418 * # repeated 0 or more times (0 means me)
423 : ( $ENV{HOME} || $ENV{LOGDIR} )
426 =head2 How come when I open a file read-write it wipes it out?
427 X<clobber> X<read-write> X<clobbering> X<truncate> X<truncating>
429 Because you're using something like this, which truncates the file and
430 I<then> gives you read-write access:
432 open(FH, "+> /path/name"); # WRONG (almost always)
434 Whoops. You should instead use this, which will fail if the file
437 open(FH, "+< /path/name"); # open for update
439 Using ">" always clobbers or creates. Using "<" never does
440 either. The "+" doesn't change this.
442 Here are examples of many kinds of file opens. Those using sysopen()
447 To open file for reading:
449 open(FH, "< $path") || die $!;
450 sysopen(FH, $path, O_RDONLY) || die $!;
452 To open file for writing, create new file if needed or else truncate old file:
454 open(FH, "> $path") || die $!;
455 sysopen(FH, $path, O_WRONLY|O_TRUNC|O_CREAT) || die $!;
456 sysopen(FH, $path, O_WRONLY|O_TRUNC|O_CREAT, 0666) || die $!;
458 To open file for writing, create new file, file must not exist:
460 sysopen(FH, $path, O_WRONLY|O_EXCL|O_CREAT) || die $!;
461 sysopen(FH, $path, O_WRONLY|O_EXCL|O_CREAT, 0666) || die $!;
463 To open file for appending, create if necessary:
465 open(FH, ">> $path") || die $!;
466 sysopen(FH, $path, O_WRONLY|O_APPEND|O_CREAT) || die $!;
467 sysopen(FH, $path, O_WRONLY|O_APPEND|O_CREAT, 0666) || die $!;
469 To open file for appending, file must exist:
471 sysopen(FH, $path, O_WRONLY|O_APPEND) || die $!;
473 To open file for update, file must exist:
475 open(FH, "+< $path") || die $!;
476 sysopen(FH, $path, O_RDWR) || die $!;
478 To open file for update, create file if necessary:
480 sysopen(FH, $path, O_RDWR|O_CREAT) || die $!;
481 sysopen(FH, $path, O_RDWR|O_CREAT, 0666) || die $!;
483 To open file for update, file must not exist:
485 sysopen(FH, $path, O_RDWR|O_EXCL|O_CREAT) || die $!;
486 sysopen(FH, $path, O_RDWR|O_EXCL|O_CREAT, 0666) || die $!;
488 To open a file without blocking, creating if necessary:
490 sysopen(FH, "/foo/somefile", O_WRONLY|O_NDELAY|O_CREAT)
491 or die "can't open /foo/somefile: $!":
493 Be warned that neither creation nor deletion of files is guaranteed to
494 be an atomic operation over NFS. That is, two processes might both
495 successfully create or unlink the same file! Therefore O_EXCL
496 isn't as exclusive as you might wish.
498 See also the new L<perlopentut> if you have it (new for 5.6).
500 =head2 Why do I sometimes get an "Argument list too long" when I use E<lt>*E<gt>?
501 X<argument list too long>
503 The C<< <> >> operator performs a globbing operation (see above).
504 In Perl versions earlier than v5.6.0, the internal glob() operator forks
505 csh(1) to do the actual glob expansion, but
506 csh can't handle more than 127 items and so gives the error message
507 C<Argument list too long>. People who installed tcsh as csh won't
508 have this problem, but their users may be surprised by it.
510 To get around this, either upgrade to Perl v5.6.0 or later, do the glob
511 yourself with readdir() and patterns, or use a module like File::KGlob,
512 one that doesn't use the shell to do globbing.
514 =head2 Is there a leak/bug in glob()?
517 Due to the current implementation on some operating systems, when you
518 use the glob() function or its angle-bracket alias in a scalar
519 context, you may cause a memory leak and/or unpredictable behavior. It's
520 best therefore to use glob() only in list context.
522 =head2 How can I open a file with a leading ">" or trailing blanks?
523 X<filename, special characters>
525 (contributed by Brian McCauley)
527 The special two argument form of Perl's open() function ignores
528 trailing blanks in filenames and infers the mode from certain leading
529 characters (or a trailing "|"). In older versions of Perl this was the
530 only version of open() and so it is prevalent in old code and books.
532 Unless you have a particular reason to use the two argument form you
533 should use the three argument form of open() which does not treat any
534 charcters in the filename as special.
536 open FILE, "<", " file "; # filename is " file "
537 open FILE, ">", ">file"; # filename is ">file"
539 =head2 How can I reliably rename a file?
540 X<rename> X<mv> X<move> X<file, rename> X<ren>
542 If your operating system supports a proper mv(1) utility or its
543 functional equivalent, this works:
545 rename($old, $new) or system("mv", $old, $new);
547 It may be more portable to use the File::Copy module instead.
548 You just copy to the new file to the new name (checking return
549 values), then delete the old one. This isn't really the same
550 semantically as a rename(), which preserves meta-information like
551 permissions, timestamps, inode info, etc.
553 Newer versions of File::Copy export a move() function.
555 =head2 How can I lock a file?
556 X<lock> X<file, lock> X<flock>
558 Perl's builtin flock() function (see L<perlfunc> for details) will call
559 flock(2) if that exists, fcntl(2) if it doesn't (on perl version 5.004 and
560 later), and lockf(3) if neither of the two previous system calls exists.
561 On some systems, it may even use a different form of native locking.
562 Here are some gotchas with Perl's flock():
568 Produces a fatal error if none of the three system calls (or their
569 close equivalent) exists.
573 lockf(3) does not provide shared locking, and requires that the
574 filehandle be open for writing (or appending, or read/writing).
578 Some versions of flock() can't lock files over a network (e.g. on NFS file
579 systems), so you'd need to force the use of fcntl(2) when you build Perl.
580 But even this is dubious at best. See the flock entry of L<perlfunc>
581 and the F<INSTALL> file in the source distribution for information on
582 building Perl to do this.
584 Two potentially non-obvious but traditional flock semantics are that
585 it waits indefinitely until the lock is granted, and that its locks are
586 I<merely advisory>. Such discretionary locks are more flexible, but
587 offer fewer guarantees. This means that files locked with flock() may
588 be modified by programs that do not also use flock(). Cars that stop
589 for red lights get on well with each other, but not with cars that don't
590 stop for red lights. See the perlport manpage, your port's specific
591 documentation, or your system-specific local manpages for details. It's
592 best to assume traditional behavior if you're writing portable programs.
593 (If you're not, you should as always feel perfectly free to write
594 for your own system's idiosyncrasies (sometimes called "features").
595 Slavish adherence to portability concerns shouldn't get in the way of
596 your getting your job done.)
598 For more information on file locking, see also
599 L<perlopentut/"File Locking"> if you have it (new for 5.6).
603 =head2 Why can't I just open(FH, "E<gt>file.lock")?
604 X<lock, lockfile race condition>
606 A common bit of code B<NOT TO USE> is this:
608 sleep(3) while -e "file.lock"; # PLEASE DO NOT USE
609 open(LCK, "> file.lock"); # THIS BROKEN CODE
611 This is a classic race condition: you take two steps to do something
612 which must be done in one. That's why computer hardware provides an
613 atomic test-and-set instruction. In theory, this "ought" to work:
615 sysopen(FH, "file.lock", O_WRONLY|O_EXCL|O_CREAT)
616 or die "can't open file.lock: $!";
618 except that lamentably, file creation (and deletion) is not atomic
619 over NFS, so this won't work (at least, not every time) over the net.
620 Various schemes involving link() have been suggested, but
621 these tend to involve busy-wait, which is also subdesirable.
623 =head2 I still don't get locking. I just want to increment the number in the file. How can I do this?
624 X<counter> X<file, counter>
626 Didn't anyone ever tell you web-page hit counters were useless?
627 They don't count number of hits, they're a waste of time, and they serve
628 only to stroke the writer's vanity. It's better to pick a random number;
629 they're more realistic.
631 Anyway, this is what you can do if you can't help yourself.
633 use Fcntl qw(:DEFAULT :flock);
634 sysopen(FH, "numfile", O_RDWR|O_CREAT) or die "can't open numfile: $!";
635 flock(FH, LOCK_EX) or die "can't flock numfile: $!";
637 seek(FH, 0, 0) or die "can't rewind numfile: $!";
638 truncate(FH, 0) or die "can't truncate numfile: $!";
639 (print FH $num+1, "\n") or die "can't write numfile: $!";
640 close FH or die "can't close numfile: $!";
642 Here's a much better web-page hit counter:
644 $hits = int( (time() - 850_000_000) / rand(1_000) );
646 If the count doesn't impress your friends, then the code might. :-)
648 =head2 All I want to do is append a small amount of text to the end of a file. Do I still have to use locking?
649 X<append> X<file, append>
651 If you are on a system that correctly implements flock() and you use the
652 example appending code from "perldoc -f flock" everything will be OK
653 even if the OS you are on doesn't implement append mode correctly (if
654 such a system exists.) So if you are happy to restrict yourself to OSs
655 that implement flock() (and that's not really much of a restriction)
656 then that is what you should do.
658 If you know you are only going to use a system that does correctly
659 implement appending (i.e. not Win32) then you can omit the seek() from
662 If you know you are only writing code to run on an OS and filesystem that
663 does implement append mode correctly (a local filesystem on a modern
664 Unix for example), and you keep the file in block-buffered mode and you
665 write less than one buffer-full of output between each manual flushing
666 of the buffer then each bufferload is almost guaranteed to be written to
667 the end of the file in one chunk without getting intermingled with
668 anyone else's output. You can also use the syswrite() function which is
669 simply a wrapper around your systems write(2) system call.
671 There is still a small theoretical chance that a signal will interrupt
672 the system level write() operation before completion. There is also a
673 possibility that some STDIO implementations may call multiple system
674 level write()s even if the buffer was empty to start. There may be some
675 systems where this probability is reduced to zero.
677 =head2 How do I randomly update a binary file?
678 X<file, binary patch>
680 If you're just trying to patch a binary, in many cases something as
681 simple as this works:
683 perl -i -pe 's{window manager}{window mangler}g' /usr/bin/emacs
685 However, if you have fixed sized records, then you might do something more
688 $RECSIZE = 220; # size of record, in bytes
689 $recno = 37; # which record to update
690 open(FH, "+<somewhere") || die "can't update somewhere: $!";
691 seek(FH, $recno * $RECSIZE, 0);
692 read(FH, $record, $RECSIZE) == $RECSIZE || die "can't read record $recno: $!";
694 seek(FH, -$RECSIZE, 1);
698 Locking and error checking are left as an exercise for the reader.
699 Don't forget them or you'll be quite sorry.
701 =head2 How do I get a file's timestamp in perl?
702 X<timestamp> X<file, timestamp>
704 If you want to retrieve the time at which the file was last
705 read, written, or had its meta-data (owner, etc) changed,
706 you use the B<-A>, B<-M>, or B<-C> file test operations as
707 documented in L<perlfunc>. These retrieve the age of the
708 file (measured against the start-time of your program) in
709 days as a floating point number. Some platforms may not have
710 all of these times. See L<perlport> for details. To
711 retrieve the "raw" time in seconds since the epoch, you
712 would call the stat function, then use localtime(),
713 gmtime(), or POSIX::strftime() to convert this into
718 $write_secs = (stat($file))[9];
719 printf "file %s updated at %s\n", $file,
720 scalar localtime($write_secs);
722 If you prefer something more legible, use the File::stat module
723 (part of the standard distribution in version 5.004 and later):
725 # error checking left as an exercise for reader.
728 $date_string = ctime(stat($file)->mtime);
729 print "file $file updated at $date_string\n";
731 The POSIX::strftime() approach has the benefit of being,
732 in theory, independent of the current locale. See L<perllocale>
735 =head2 How do I set a file's timestamp in perl?
736 X<timestamp> X<file, timestamp>
738 You use the utime() function documented in L<perlfunc/utime>.
739 By way of example, here's a little program that copies the
740 read and write times from its first argument to all the rest
744 die "usage: cptimes timestamp_file other_files ...\n";
747 ($atime, $mtime) = (stat($timestamp))[8,9];
748 utime $atime, $mtime, @ARGV;
750 Error checking is, as usual, left as an exercise for the reader.
752 The perldoc for utime also has an example that has the same
753 effect as touch(1) on files that I<already exist>.
755 Certain file systems have a limited ability to store the times
756 on a file at the expected level of precision. For example, the
757 FAT and HPFS filesystem are unable to create dates on files with
758 a finer granularity than two seconds. This is a limitation of
759 the filesystems, not of utime().
761 =head2 How do I print to more than one file at once?
762 X<print, to multiple files>
764 To connect one filehandle to several output filehandles,
765 you can use the IO::Tee or Tie::FileHandle::Multiplex modules.
767 If you only have to do this once, you can print individually
770 for $fh (FH1, FH2, FH3) { print $fh "whatever\n" }
772 =head2 How can I read in an entire file all at once?
773 X<slurp> X<file, slurping>
775 You can use the File::Slurp module to do it in one step.
779 $all_of_it = read_file($filename); # entire file in scalar
780 @all_lines = read_file($filename); # one line perl element
782 The customary Perl approach for processing all the lines in a file is to
783 do so one line at a time:
785 open (INPUT, $file) || die "can't open $file: $!";
788 # do something with $_
790 close(INPUT) || die "can't close $file: $!";
792 This is tremendously more efficient than reading the entire file into
793 memory as an array of lines and then processing it one element at a time,
794 which is often--if not almost always--the wrong approach. Whenever
795 you see someone do this:
799 you should think long and hard about why you need everything loaded at
800 once. It's just not a scalable solution. You might also find it more
801 fun to use the standard Tie::File module, or the DB_File module's
802 $DB_RECNO bindings, which allow you to tie an array to a file so that
803 accessing an element the array actually accesses the corresponding
806 You can read the entire filehandle contents into a scalar.
810 open (INPUT, $file) || die "can't open $file: $!";
814 That temporarily undefs your record separator, and will automatically
815 close the file at block exit. If the file is already open, just use this:
817 $var = do { local $/; <INPUT> };
819 For ordinary files you can also use the read function.
821 read( INPUT, $var, -s INPUT );
823 The third argument tests the byte size of the data on the INPUT filehandle
824 and reads that many bytes into the buffer $var.
826 =head2 How can I read in a file by paragraphs?
827 X<file, reading by paragraphs>
829 Use the C<$/> variable (see L<perlvar> for details). You can either
830 set it to C<""> to eliminate empty paragraphs (C<"abc\n\n\n\ndef">,
831 for instance, gets treated as two paragraphs and not three), or
832 C<"\n\n"> to accept empty paragraphs.
834 Note that a blank line must have no blanks in it. Thus
835 S<C<"fred\n \nstuff\n\n">> is one paragraph, but C<"fred\n\nstuff\n\n"> is two.
837 =head2 How can I read a single character from a file? From the keyboard?
838 X<getc> X<file, reading one character at a time>
840 You can use the builtin C<getc()> function for most filehandles, but
841 it won't (easily) work on a terminal device. For STDIN, either use
842 the Term::ReadKey module from CPAN or use the sample code in
845 If your system supports the portable operating system programming
846 interface (POSIX), you can use the following code, which you'll note
847 turns off echo processing as well.
861 use POSIX qw(:termios_h);
863 my ($term, $oterm, $echo, $noecho, $fd_stdin);
865 $fd_stdin = fileno(STDIN);
867 $term = POSIX::Termios->new();
868 $term->getattr($fd_stdin);
869 $oterm = $term->getlflag();
871 $echo = ECHO | ECHOK | ICANON;
872 $noecho = $oterm & ~$echo;
875 $term->setlflag($noecho);
876 $term->setcc(VTIME, 1);
877 $term->setattr($fd_stdin, TCSANOW);
881 $term->setlflag($oterm);
882 $term->setcc(VTIME, 0);
883 $term->setattr($fd_stdin, TCSANOW);
889 sysread(STDIN, $key, 1);
898 The Term::ReadKey module from CPAN may be easier to use. Recent versions
899 include also support for non-portable systems as well.
902 open(TTY, "</dev/tty");
903 print "Gimme a char: ";
905 $key = ReadKey 0, *TTY;
907 printf "\nYou said %s, char number %03d\n",
910 =head2 How can I tell whether there's a character waiting on a filehandle?
912 The very first thing you should do is look into getting the Term::ReadKey
913 extension from CPAN. As we mentioned earlier, it now even has limited
914 support for non-portable (read: not open systems, closed, proprietary,
915 not POSIX, not Unix, etc) systems.
917 You should also check out the Frequently Asked Questions list in
918 comp.unix.* for things like this: the answer is essentially the same.
919 It's very system dependent. Here's one solution that works on BSD
924 vec($rin, fileno(STDIN), 1) = 1;
925 return $nfd = select($rin,undef,undef,0);
928 If you want to find out how many characters are waiting, there's
929 also the FIONREAD ioctl call to be looked at. The I<h2ph> tool that
930 comes with Perl tries to convert C include files to Perl code, which
931 can be C<require>d. FIONREAD ends up defined as a function in the
932 I<sys/ioctl.ph> file:
934 require 'sys/ioctl.ph';
936 $size = pack("L", 0);
937 ioctl(FH, FIONREAD(), $size) or die "Couldn't call ioctl: $!\n";
938 $size = unpack("L", $size);
940 If I<h2ph> wasn't installed or doesn't work for you, you can
941 I<grep> the include files by hand:
943 % grep FIONREAD /usr/include/*/*
944 /usr/include/asm/ioctls.h:#define FIONREAD 0x541B
946 Or write a small C program using the editor of champions:
949 #include <sys/ioctl.h>
951 printf("%#08x\n", FIONREAD);
954 % cc -o fionread fionread.c
958 And then hard code it, leaving porting as an exercise to your successor.
960 $FIONREAD = 0x4004667f; # XXX: opsys dependent
962 $size = pack("L", 0);
963 ioctl(FH, $FIONREAD, $size) or die "Couldn't call ioctl: $!\n";
964 $size = unpack("L", $size);
966 FIONREAD requires a filehandle connected to a stream, meaning that sockets,
967 pipes, and tty devices work, but I<not> files.
969 =head2 How do I do a C<tail -f> in perl?
976 The statement C<seek(GWFILE, 0, 1)> doesn't change the current position,
977 but it does clear the end-of-file condition on the handle, so that the
978 next <GWFILE> makes Perl try again to read something.
980 If that doesn't work (it relies on features of your stdio implementation),
981 then you need something more like this:
984 for ($curpos = tell(GWFILE); <GWFILE>; $curpos = tell(GWFILE)) {
985 # search for some stuff and put it into files
988 seek(GWFILE, $curpos, 0); # seek to where we had been
991 If this still doesn't work, look into the POSIX module. POSIX defines
992 the clearerr() method, which can remove the end of file condition on a
993 filehandle. The method: read until end of file, clearerr(), read some
994 more. Lather, rinse, repeat.
996 There's also a File::Tail module from CPAN.
998 =head2 How do I dup() a filehandle in Perl?
1001 If you check L<perlfunc/open>, you'll see that several of the ways
1002 to call open() should do the trick. For example:
1004 open(LOG, ">>/foo/logfile");
1005 open(STDERR, ">&LOG");
1007 Or even with a literal numeric descriptor:
1009 $fd = $ENV{MHCONTEXTFD};
1010 open(MHCONTEXT, "<&=$fd"); # like fdopen(3S)
1012 Note that "<&STDIN" makes a copy, but "<&=STDIN" make
1013 an alias. That means if you close an aliased handle, all
1014 aliases become inaccessible. This is not true with
1017 Error checking, as always, has been left as an exercise for the reader.
1019 =head2 How do I close a file descriptor by number?
1020 X<file, closing file descriptors>
1022 This should rarely be necessary, as the Perl close() function is to be
1023 used for things that Perl opened itself, even if it was a dup of a
1024 numeric descriptor as with MHCONTEXT above. But if you really have
1025 to, you may be able to do this:
1027 require 'sys/syscall.ph';
1028 $rc = syscall(&SYS_close, $fd + 0); # must force numeric
1029 die "can't sysclose $fd: $!" unless $rc == -1;
1031 Or, just use the fdopen(3S) feature of open():
1035 open F, "<&=$fd" or die "Cannot reopen fd=$fd: $!";
1039 =head2 Why can't I use "C:\temp\foo" in DOS paths? Why doesn't `C:\temp\foo.exe` work?
1040 X<filename, DOS issues>
1042 Whoops! You just put a tab and a formfeed into that filename!
1043 Remember that within double quoted strings ("like\this"), the
1044 backslash is an escape character. The full list of these is in
1045 L<perlop/Quote and Quote-like Operators>. Unsurprisingly, you don't
1046 have a file called "c:(tab)emp(formfeed)oo" or
1047 "c:(tab)emp(formfeed)oo.exe" on your legacy DOS filesystem.
1049 Either single-quote your strings, or (preferably) use forward slashes.
1050 Since all DOS and Windows versions since something like MS-DOS 2.0 or so
1051 have treated C</> and C<\> the same in a path, you might as well use the
1052 one that doesn't clash with Perl--or the POSIX shell, ANSI C and C++,
1053 awk, Tcl, Java, or Python, just to mention a few. POSIX paths
1054 are more portable, too.
1056 =head2 Why doesn't glob("*.*") get all the files?
1059 Because even on non-Unix ports, Perl's glob function follows standard
1060 Unix globbing semantics. You'll need C<glob("*")> to get all (non-hidden)
1061 files. This makes glob() portable even to legacy systems. Your
1062 port may include proprietary globbing functions as well. Check its
1063 documentation for details.
1065 =head2 Why does Perl let me delete read-only files? Why does C<-i> clobber protected files? Isn't this a bug in Perl?
1067 This is elaborately and painstakingly described in the
1068 F<file-dir-perms> article in the "Far More Than You Ever Wanted To
1069 Know" collection in http://www.cpan.org/misc/olddoc/FMTEYEWTK.tgz .
1071 The executive summary: learn how your filesystem works. The
1072 permissions on a file say what can happen to the data in that file.
1073 The permissions on a directory say what can happen to the list of
1074 files in that directory. If you delete a file, you're removing its
1075 name from the directory (so the operation depends on the permissions
1076 of the directory, not of the file). If you try to write to the file,
1077 the permissions of the file govern whether you're allowed to.
1079 =head2 How do I select a random line from a file?
1080 X<file, selecting a random line>
1082 Here's an algorithm from the Camel Book:
1085 rand($.) < 1 && ($line = $_) while <>;
1087 This has a significant advantage in space over reading the whole file
1088 in. You can find a proof of this method in I<The Art of Computer
1089 Programming>, Volume 2, Section 3.4.2, by Donald E. Knuth.
1091 You can use the File::Random module which provides a function
1094 use File::Random qw/random_line/;
1095 my $line = random_line($filename);
1097 Another way is to use the Tie::File module, which treats the entire
1098 file as an array. Simply access a random array element.
1100 =head2 Why do I get weird spaces when I print an array of lines?
1106 joins together the elements of C<@lines> with a space between them.
1107 If C<@lines> were C<("little", "fluffy", "clouds")> then the above
1108 statement would print
1110 little fluffy clouds
1112 but if each element of C<@lines> was a line of text, ending a newline
1113 character C<("little\n", "fluffy\n", "clouds\n")> then it would print:
1119 If your array contains lines, just print them:
1125 Revision: $Revision: 3606 $
1127 Date: $Date: 2006-03-06 12:05:47 +0100 (lun, 06 mar 2006) $
1129 See L<perlfaq> for source control details and availability.
1131 =head1 AUTHOR AND COPYRIGHT
1133 Copyright (c) 1997-2006 Tom Christiansen, Nathan Torkington, and
1134 other authors as noted. All rights reserved.
1136 This documentation is free; you can redistribute it and/or modify it
1137 under the same terms as Perl itself.
1139 Irrespective of its distribution, all code examples here are in the public
1140 domain. You are permitted and encouraged to use this code and any
1141 derivatives thereof in your own programs for fun or for profit as you
1142 see fit. A simple comment in the code giving credit to the FAQ would
1143 be courteous but is not required.